ChatGPT, Gemini and Grok all fail at an image most humans recognize instantly
JeremyNguyenPhD · x · 2026-09-27
User @ghostcartoonist shared an image that most people can instantly interpret, but ChatGPT, Gemini, and Grok all failed to describe what's happening in it. Jeremy Nguyen highlighted it as an interesting challenge for AI vision models: an easy task for human intuition that stump mainstream multimodal models — a notable observation on the gap between current visual understanding and human perception.
More from Fun
- Developer dreams his Opus 5.5 agent ran idle all night, calls it a FOMO nightmare — ThePeterMick · 2026-09-27
- Opus-coded betta fish desktop wallpaper chases your cursor, open-sourced as Deskworlds — chaseleantj · 2026-09-27
- Gary Marcus shares alignment meme: misaligned arm and cortex team up to steal cookies — GaryMarcus · 2026-09-27
- Stochastic Strawmen: Can LLM Critics Hold a Consistent Argument for More Than a Year? — Zergylord · 2026-09-27
- Sony Retro Display Project's Enclosure Takes Shape After Repeated Refinement — Jasonio · 2026-09-27
- Anti-AI protesters rush stage at Data After Dark party in DC, drowned out by 'USA!' chants — StewartalsopIII · 2026-09-27